Do you think this good answer should be edited to reflect how --with-new-pkgs is implied when one uses apt rather than apt-get to perform the upgrade action?
@EliahKagan maybe I just can't get over my discomfort with the first sentence... but the alternative suggested at the beginning is not quite the one we are talking about at the end... maybe the style needs to be more integrated, but I don't want to dismantle the author's original post
@Zanna Yeah, I see two remaining problems. The first problem pertains to gvfs-trash vs trash: it's unclear at first which alterative gives a trash command that be called from find, as well as how to decide between them.
The second problem is that there is a nonrecursive approach and a recursive approach, and they're both valuable. Often the nonrecursive approach is preferable or even the only correct way (when one know what's immediately in . but not what naming schemes have been used in subdirectories). They both deserve to be explained and to be presented with commands that show what files they will trash as well as commands that actually perform the trashing.
@Zanna Usually canned comments on answers that look like that are wrong, but I think answer is a success report and attempt to reply to an existing comment on that other answer.
Maybe it is giving garbage values on Ubuntu but not another OS. But they have not said that. Also, if so, that would sort of be a gray area, because it would only be by total chance that undefined behavior would appear to work on another OS, and which OS is used is often not even the most important factor that determines that.
@Zanna Well, the phrase "generic programming" has other meanings; maybe that's the reason.
Has the Update Manager ever used sudo? I think it may once have been run with gksu update-manager, causing sudo to be used to start it, but has the program itself ever used sudo?
Should that answer be edited to give a different example? To mention Polkit?